aiagents4pharma 0.0.0__py3-none-any.whl

aiagents4pharma/talk2knowledgegraphs/utils/enrichments/reactome_pathways.py

@@ -0,0 +1,71 @@
+#!/usr/bin/env python3
+
+"""
+Enrichment class for enriching Reactome pathways with textual descriptions
+"""
+
+import logging
+
+import hydra
+import requests
+
+from .enrichments import Enrichments
+
+# Initialize logger
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
+
+class EnrichmentWithReactome(Enrichments):
+    """
+    Enrichment class using Reactome pathways
+    """
+
+    def enrich_documents(self, texts: list[str]) -> list[str]:
+        """
+        Enrich a list of input Reactome pathways
+
+        Args:
+            texts: The list of Reactome pathways to be enriched.
+
+        Returns:
+            The list of enriched descriptions
+        """
+
+        reactome_pathways_ids = texts
+
+        logger.log(logging.INFO, "Load Hydra configuration for reactome enrichment")
+        with hydra.initialize(version_base=None, config_path="../../configs"):
+            cfg = hydra.compose(
+                config_name="config",
+                overrides=["utils/enrichments/reactome_pathways=default"],
+            )
+            cfg = cfg.utils.enrichments.reactome_pathways
+
+        descriptions = []
+        for reactome_pathway_id in reactome_pathways_ids:
+            r = requests.get(
+                cfg.base_url + reactome_pathway_id + "/summation",
+                headers={"Accept": "text/plain"},
+                timeout=cfg.timeout,
+            )
+            # if the response is not ok
+            if not r.ok:
+                descriptions.append(None)
+                continue
+            response_body = r.text
+            # if the response is ok
+            descriptions.append(response_body.split("\t")[1])
+        return descriptions
+
+    def enrich_documents_with_rag(self, texts, docs):
+        """
+        Enrich a list of input Reactome pathways
+
+        Args:
+            texts: The list of Reactome pathways to be enriched.
+
+        Returns:
+            The list of enriched descriptions
+        """
+        return self.enrich_documents(texts)

aiagents4pharma/talk2knowledgegraphs/utils/enrichments/uniprot_proteins.py

@@ -0,0 +1,98 @@
+#!/usr/bin/env python3
+
+"""
+Enrichment class for enriching Gene names with their function and sequence using UniProt.
+"""
+
+import json
+import logging
+
+import hydra
+import requests
+
+from .enrichments import Enrichments
+
+# Initialize logger
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
+
+class EnrichmentWithUniProt(Enrichments):
+    """
+    Enrichment class using UniProt
+    """
+
+    def enrich_documents(self, texts: list[str]) -> list[str]:
+        """
+        Enrich a list of input UniProt gene names with their function and sequence.
+
+        Args:
+            texts: The list of gene names to be enriched.
+
+        Returns:
+            The list of enriched functions and sequences
+        """
+
+        enriched_gene_names = texts
+
+        logger.log(
+            logging.INFO,
+            "Load Hydra configuration for Gene enrichment with description and sequence.",
+        )
+        with hydra.initialize(version_base=None, config_path="../../configs"):
+            cfg = hydra.compose(
+                config_name="config",
+                overrides=["utils/enrichments/uniprot_proteins=default"],
+            )
+            cfg = cfg.utils.enrichments.uniprot_proteins
+
+        descriptions = []
+        sequences = []
+        for gene in enriched_gene_names:
+            params = {
+                "reviewed": cfg.reviewed,
+                "isoform": cfg.isoform,
+                "exact_gene": gene,
+                "organism": cfg.organism,
+                # You can get the list of all available organisms here:
+                # https://www.uniprot.org/help/taxonomy
+            }
+
+            r = requests.get(
+                cfg.uniprot_url,
+                headers={"Accept": "application/json"},
+                params=params,
+                timeout=cfg.timeout,
+            )
+            # if the response is not ok
+            if not r.ok:
+                descriptions.append(None)
+                sequences.append(None)
+                continue
+            response_body = json.loads(r.text)
+            # if the response body is empty
+            if not response_body:
+                descriptions.append(None)
+                sequences.append(None)
+                continue
+            description = ""
+            for comment in response_body[0]["comments"]:
+                if comment["type"] == "FUNCTION":
+                    for value in comment["text"]:
+                        description += value["value"]
+            sequence = response_body[0]["sequence"]["sequence"]
+            descriptions.append(description)
+            sequences.append(sequence)
+        return descriptions, sequences
+
+    def enrich_documents_with_rag(self, texts, docs):
+        """
+        Enrich a list of input UniProt gene names with their function and sequence.
+
+        Args:
+            texts: The list of gene names to be enriched.
+
+        Returns:
+            The list of enriched functions and sequences
+        """
+        return self.enrich_documents(texts)

aiagents4pharma/talk2knowledgegraphs/utils/extractions/__init__.py

@@ -0,0 +1,5 @@
+"""
+This file is used to import all the models in the package.
+"""
+
+from . import milvus_multimodal_pcst, multimodal_pcst, pcst